1. Field of the Invention
The present invention relates to a solder terminal structure and a method for fabricating the same and, more particularly, to a solder terminal structure and a fabrication method thereof to improve the reliability of a solder terminal.
2. Description of the Related Art
A so-called flip chip bonding technology is commonly used for semiconductor device packaging. The flip-chip bonding technology enables electrical and physical connection between a semiconductor chip and a substrate such as a ceramic substrate or a circuit board by simultaneously connecting corresponding terminals of the semiconductor chip and the substrate using solder balls or solder columns.
FIG. 1 is a cross-sectional view of a solder bump assembly to illustrate the connections between a semiconductor chip and a substrate using a conventional flip-chip bonding technology.
Referring to FIG. 1, solder balls 45 or solder bumps 40 electrically connect electrode (bond) pads 20 on a semiconductor device 30 and electrode pads 11 on a substrate 10. The solder balls 45 or solder bumps 40 are formed on one side or both sides of the semiconductor device 30 or substrate 10 and are connected to the corresponding electrode pads 20, 11 exposed through insulating films 50 that cover the semiconductor device 30 and the substrate 10.
A reference numeral “51” indicates an epoxy resin filled between the semiconductor device 30 and the substrate 10 to protect the semiconductor device 30 and substrate 10 from harmful environmental conditions such as electrical and physical impact.
To precisely control the volume and composition of the solder, the solder terminals are manufactured by selectively depositing or plating the solder using a mask formed of a metal or photoresist.
The solder bumps 40 formed on the semiconductor device 30 are generally separated into two areas of a solder terminal metal layer and a connection layer.
On the other hand, because metals such as aluminum, chrome, molybdenum or tungsten employed for wiring of the semiconductor device 30 do not flow or react well, the adhesion between the solder and the adjacent layers has been somewhat problematic.
FIG. 2A is a plan view showing the electrode pads 20 formed on the surface of the semiconductor device 30 and the solder bumps 40 connected to the electrode pads 20. Referring to FIG. 2A, the plurality of electrode pads 20 formed on the surface of the semiconductor device 30 are isolated from each other and exposed through the insulating layer 50. And a solder bump 40 is attached to the electrode pads 20 through the solder terminal (not shown). The position, number and size of the solder terminals are determined by a design rule of the semiconductor device 30 and the substrate 10, module reliability and the solder terminal forming process.
Conventionally, the total number of mutual connections used in a Very Large Scale Integration (VLSI) is about 1,000 and is expected to increase further. In line with current demands for high-speed and high-performance semiconductor devices, it is required that the electrode pads 20 of the semiconductor 30 have very high thermal and mechanical reliability.
FIG. 2B is an exemplary view showing the electrode pad 20 and a solder bump terminal on the semiconductor device 30 according to the prior art. With reference to FIG. 2B, the semiconductor device 30 has at least one patterned wiring and a predetermined area of the wiring is exposed through the insulating layer 50 on the surface of the semiconductor device 30 to form the electrode pad 20. An adhesion metal layer 60 is formed on the insulating layer 50 and the electrode pad 20.
Also, a solder bonding layer 80 formed of a material for soldering forms a metallurgical bond by reacting with solder when the solder is heated. An intermediate layer 70 is formed between the adhesion metal layer 60 and solder bonding layer 80 to increase the bonding strength between the adhesion metal layer 60 and solder bonding layer 80, if necessary.
On the other hand, the solder bonding layer 80 reacts with at least one component of the solder bump 40 when the solder is melted, and forms an inter-metallic compound. When the solder is repeatedly melted numerous times, a Cu—Sn inter-metallic compound (not shown) grows thick and the metallurgical microstructure becomes coarse and discontinuous.
Therefore, in the prior art, if such a reaction continues, all of the solder bonding layer 80 can be changed into an intermetallic compound and a bonding strength of the solder bump 40 and adhesion metal layer 60 undesirably declines.
To prevent such a reduction in bonding strength of the solder bump 40 and the adhesion metal layer 60, the intermediate layer 70 is inserted between the solder bump 40 and adhesion metal layer 70 as discussed above. Conventionally, Cr—Cu is employed to form the intermediate layer 70.
With the conventional intermediate layer formed of, e.g., Cr—Cu, however, the Cu component in the Cr—Cu reacts with Sn in the solder and is thus removed, thus reducing the adhesion strength of the solder, especially, the strength of the solder where the solder contacts the other adjacent layers.
Further, with the prior art structure, there can be harmful reaction or interdiffusion between the electrode pad and the solder bonding layer, or between the adhesion layer and the solder bonding layer, which significantly degrades the device characteristics.
In addition, during the reliability test of the semiconductor devices, a probe mark may be left on the electrode (bond) pad. Because the area of the electrode pad where the probe mark is left is relatively thin, if a solder terminal of another material is directly mounted thereon, the electrode pad may be damaged by stress and chemical reaction between different materials.